Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    matrixbot
    @matrixbot
    shrit When I use cout I can not see the difference between row and columns, all of them are streamed to output.
    rcurtin on Freenode I guess because there are too many columns and they wrap?
    shrit Actually there is about 10K rows and three columns
    rcurtin on Freenode oh, maybe print with .t()?
    rcurtin on Freenode but I'm not totally sure what you mean, maybe paste the output?
    matrixbot
    @matrixbot
    shrit Exactly it is much easier to read when using transposing the matrix
    shrit @rcurtin, another question considering the meaning of the word ┬┤coordinate┬┤ in ensmallen. If I use ensmallen optimizer to train a neural network the word coordinate means the weights of the netowrk, am I right?
    matrixbot
    @matrixbot
    rcurtin on Freenode shrit: right, coordinate == weight of a network
    shrit Exactly, So If I want to store the best coordinate, I will have to use the callback ens::StoreBestCoordinate<> and then these coordinates are stored to a specific matrix, I want just to ask how to be sure that the Best coordinates are saved into the model.bin at the end?
    matrixbot
    @matrixbot
    zoq on Freenode shrit: Before saving, call model.Parameters() = bestParameter;
    shrit I will do a pull request and update an existing example, this one worth to be noted down, many thanks :+1
    matrixbot
    @matrixbot
    travis-ci on Freenode shrit/examples#99 (best_coordinates - 2bb362b : Omar Shrit): The build passed.
    travis-ci on Freenode Build details : https://travis-ci.com/shrit/examples/builds/190524716
    matrixbot
    @matrixbot
    travis-ci on Freenode shrit/examples#101 (fix_mnist_simple_bug - 238aff2 : shrit): The build passed.
    travis-ci on Freenode Build details : https://travis-ci.com/shrit/examples/builds/190569693
    matrixbot
    @matrixbot
    rcurtin on Freenode I thought it's about time for another release, so I made a milestone and put in a few PRs that look pretty close:
    rcurtin on Freenode I didn't add any of the catch or WeightSize refactorings, since those aren't actually user visible (so if they get in, great, if not, it shouldn't matter for users)
    rcurtin on Freenode if anyone else has any PRs that are "just about" done and can be merged in the next couple of days, feel free to point them out or just add them to the milestone! :)
    matrixbot
    @matrixbot
    abernauer rcurtin: Dumb git question, but I sort of messed up my worflow for that PR for the r bindings example. I wrote the code in a seperate repo I can just merge those commits into a branch on my fork of the examples repo correct?
    matrixbot
    @matrixbot
    rcurtin on Freenode if it's a different repository entirely, then you may not be able to merge those commits, but you could always just copy the files over and add them to your fork
    abernauer Yeah the latter is easy enouth
    abernauer *enough
    matrixbot
    @matrixbot
    shrit Just to share one test I have done. I have one piece of code that include mlpack core (To use armadillo), and another piece of code that use FFN code to create a model, the first one requires 55 seconds to compile, the seconde one require 4:35 minutes to compile
    shrit The difference is only the inclusion of some layers + FFN headers, there not even ensmallen code nor optimizer
    shrit I think that boost visitors are taking a considerable amount of time, there is no way that template usage in FFN code to consume such amount of time.
    Aakash kaushik
    @Aakash-kaushik
    Hey @shrit i am working to see the difference by removing the boost::visitors
    and instead using virtual inheritance, I was running a bit busy due to exams but i will start working on it.
    Aakash kaushik
    @Aakash-kaushik
    Hey BOOST_REQUIRE(x) should be translated into something like REQUIRE(x == true) ?
    Anush Kini
    @Abilityguy
    Yes @Aakash-kaushik. REQUIRE(x) should also work.
    Aakash kaushik
    @Aakash-kaushik
    @Abilityguy thanks, this seems like a more elegant solution.
    Aakash kaushik
    @Aakash-kaushik

    There is a test in fastmks_test.cpp at line 180 which fails with this expansion:

    REQUIRE( pk.Evaluate(denseset.col(i), denseset.col(j)) == 0.0 )
    with expansion:
      0.0 == 0.0

    I tried to use Approx(0.0) then it was failing with 0.0 == Approx(0.0) and even tried Approx(0.0).epsilon(1e-10) it was still failing with 0.0 == Approx(0.0)
    I don't seem to get how getting 0.0 == 0.0 fails the test ?

    matrixbot
    @matrixbot
    jeffin143 shrit : planning to migrate serialisation test to catch?
    Are you ok with it ??
    There would be some merge conflicts if I am not wrong ?? Would you be ok to handle those ??
    Aakash kaushik
    @Aakash-kaushik
    Hey @jeffin143 i can handle most of the remaining tests so if you are actively working on one that doesn't have a PR, let me know so i don't take up the same
    matrixbot
    @matrixbot
    shrit @jeffin143 I am not planning to migrate anything at this time, but I prefere that you wait until cereal is merged, it is very close to be merged at this time. You can do a pull request and do a migration of course, but we can wait for merging I think.
    matrixbot
    @matrixbot
    srinivasyadav227 hey I am new to mlpack, can anyone help me through getting started ?
    srinivasyadav227 abernauer: thankyou
    abernauer yeah np.
    Aakash kaushik
    @Aakash-kaushik
    Hi, what's the translation for BOOST_REQUIRE_EQUAL_COLLECTIONS in catch2 like i understand that i may have to run two loops for something like a vector but then how to run two loops in parallel, or is there something direct in catch2 ?
    matrixbot
    @matrixbot
    travis-ci on Freenode mlpack/ensmallen#1066 (master - 25d0d84 : Marcus Edel): The build was fixed.
    travis-ci on Freenode Build details : https://travis-ci.org/mlpack/ensmallen/builds/737204331
    Aakash kaushik
    @Aakash-kaushik
    I think i will just use indexes instead of iterators for the BOOST_REQUIRE_EQUAL_COLLECTIONS
    matrixbot
    @matrixbot
    rcurtin on Freenode Aakash kaushik (Gitter): that sounds just fine to me; really what matters most is just that the tests do the same thing in the end, even if we have to refactor them a little bit :)
    Aakash kaushik
    @Aakash-kaushik
    Thanks @rcurtin :D